Water spray detection for smart irrigation systems with Mask R-CNN and UAV footage

While the world’s population rises, demand for food grows accordingly. Smart agriculture emerges as a viable solution to increase the quality and efficiency of crops. Irrigation plays an essential role in the grade and productivity of harvests, while also being a crucial factor in the cost-effectiveness of food production. Smart irrigation uses technology to improve watering, such as the Internet of Things (IoT) applications and Machine Learning algorithms. The correct functioning of irrigation nozzles is critical to ensure that the hydration plan is deployed correctly to the crop field. This paper presents a Machine Learning algorithm that can automatically recognize water from aerial footage of irrigation systems. This automatic recognition can help in the irrigation system inspection, potentially reducing time and cost in system maintenance. Initial results show that it is possible to identify water on image frames captured by an Unmanned Aerial Vehicle (UAV) using the Mask R-CNN Neural Network. The goal is to identify malfunctioning irrigation systems that can lead to under or overwatering, compromising the irrigation plan’s correct implementation.

[1]  Juha-Pekka Soininen,et al.  Smart Water Management Platform: IoT-Based Precision Irrigation for Agriculture † , 2019, Sensors.

[2]  Abhishek Dutta,et al.  The VIA Annotation Software for Images, Audio and Video , 2019, ACM Multimedia.

[3]  L. Khot,et al.  Assessing suitability of modified center pivot irrigation systems in corn production using low altitude aerial imaging techniques , 2020, Information Processing in Agriculture.

[4]  Ross B. Girshick,et al.  Mask R-CNN , 2017, 1703.06870.

[5]  Trevor Darrell,et al.  Fully Convolutional Networks for Semantic Segmentation , 2017, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Ronald M. Summers,et al.  Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning , 2016, IEEE Transactions on Medical Imaging.

[7]  R. Nagarajan,et al.  Drone survey facilitated weeds assessment and impact on hydraulic efficiency of canals , 2018, ISH Journal of Hydraulic Engineering.

[8]  Ian J. Yule,et al.  Key Performance Indicators for Simulated Variable-Rate Irrigation of Variable Soils in Humid Regions , 2009 .

[9]  Pietro Perona,et al.  Microsoft COCO: Common Objects in Context , 2014, ECCV.

[10]  António Pereira,et al.  A Systematic Review of IoT Solutions for Smart Farming , 2020, Sensors.

[11]  Aske Plaat,et al.  On the Impact of Data Set Size in Transfer Learning Using Deep Neural Networks , 2016, IDA.

[12]  Maria Mar Alsina,et al.  Improving vineyard water use efficiency and yield with variable rate irrigation in California , 2017 .

[13]  Tania Prinsloo,et al.  The influence of drone monitoring on crop health and harvest size , 2017, 2017 1st International Conference on Next Generation Computing Applications (NextComp).

[14]  F. Sarghini,et al.  First measurements of spray deposition obtained from UAV spray application technique , 2019, 2019 IEEE International Workshop on Metrology for Agriculture and Forestry (MetroAgriFor).

[15]  Gyanendra K. Verma,et al.  Convolutional neural network: a review of models, methodologies and applications to object detection , 2019, Progress in Artificial Intelligence.

[16]  Luis S. Pereira,et al.  DEPIVOT: A model for center-pivot design and evaluation , 2012 .

[17]  Ronaldo C. Prati,et al.  Advancing IoT-Based Smart Irrigation , 2019, IEEE Internet of Things Magazine.